In this paper, we will present a lossless data compression technique called statistical the Lempel-Ziv (LZ) compression algorithm for personal digital assistants (PDAs). This compression algorithm may be viewed as a variant of the LZ77 and the contribution of this algorithm is to include the statistical properties of the source information while most of the LZ-based compression methods, such as LZ78 and LZW do not take this property into consideration. In addition, a prefix entropy-coding scheme is designed to improve the look-up table time for decoding. These prefix codes are especially suitable for PDAs in certain situations. The decoding of prefix codes works very fast by using simple logical and arithmetic operations.